I'm going to ask this question here too.. are we talking about the DNS
or are we talking about an applications use of data published in the DNS?
i see this draft in the context of the historical DNS ... it is a mapping
service, a name to an address AND an address to a name. the mapping service
is OPTIONAL ... the Internet does not now (and should NEVER) depend on this
service being available for any other service to work.
I think a key point here is that -reciprocity- is and should be highly valued
feature of DNS provisioning.
that said, if one provisions mapping in one direction, it should also be
provided in the other.
e.g.
label1 == address1 implies address1 == label1
so, if it is legal to have this construct:
label == address1
== address2
...
== addressN
then this construct should also be legal:
address == name1
== name2
...
== nameN
what we have is a mess....
label1 == address1918
address1918 == lable8
which, while technicallu legal, violates the spirit of the
lema posited above, as does;
label35 == address1918
and address1918 has no matching label.
we now return you to your SMTP coloured discussion.
--bill
_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop
